Tuning Parameters in Postgres vs. Tuning Your Queries
summary
Compares changing Postgres configuration versus fixing queries using the postgres_air sample database and live timings: with defaults the query runs ~2.5s, enabling parallelism drops to ~2.1s, raising work_mem 500MB to ~1.8s, increasing shared_buffers to 1GB reaches ~1.0s; adding an index on actual_departure gives ~0.7s and adding a booking_id index plus a query rewrite yields 60 ms, a ~10x win. Notes parameters usually buy ~10–50% while query and application changes can deliver orders-of-magnitude improvement.
AI-written summary · about
chapters (6)
- 0:00Intro
- 1:19Tuning your database - what does it mean?
- 5:28Query example
- 10:03Let's try something different!
- 11:55Understanding the role of parameters
- 15:31Application changes
related talks
18mPostgreSQL: Tuning parameters or Tuning Queries? with Henrietta Dombrovskaya
43mPostgres Performance: From Slow to Pro
33mTuning PostgreSQL to work even better
34mTuning PostgreSQL to work even better
45mProper PostgreSQL Parameters to Prevent Poor Performance
25mScaling PostgreSQL Beyond Query Optimization
04mPostgres Tuning is saved by Sovenath Shaw
27mPostgres Performance in 15 Minutes